Quick Summary:

When we think about grammar, there are two different kinds of rules that get discussed, prescriptive rules and descriptive rules. Prescriptive rules are the ones that tell you how you should speak: don't leave a preposition at the end of a sentence, don't say "no problem" when you mean "thank you", don't put "like" in between words if you're not using it as a comparison. These are rules you have to be taught, that you know consciously, and that you can break if you want (and you probably do). They're just social conventions, and there's nothing fundamental or basic about them.

Descriptive rules, on the other hand, show how people actually behave and use their language. These are the ones that actually guide your interpretations, tell you what sound combinations are okay for your language and which ones aren't, and how to build complex sentences you've never said or heard before. You might not be aware of what these rules are, or be able to explain them, but you apply them to speech all of the time subconsciously. These principles of language what makes our language work; they're what lies within our real grammar, and they're what linguists study when we do our research.



We mentioned a couple of phenomena in English that are dependent upon descriptive rules that (hopefully!) everyone can agree with. Trying to explain them in the video itself would have made things quite long, so we have them here for our extras.

The first of these was the comparison between "steek" and "sfeek". We think that "steek" is the better of these words, and that should be pretty uncontroversial. So what leads to that judgment?

One thing we can notice first is that there's quite a large difference between the number of words in English that start off with that [st] sound cluster - steak, stick, stump, star, stunt, etc. But it turns out that there are only two English words that start with [sf], sphere and sphinx. So maybe we can just think, well, just typologically, there are just so many more [st] words, they're more likely to be better, so we prefer "steek". Maybe other languages that have more [sf] words in them would like "sfik" better.

But that can't really capture the whole matter. There's a general principle for words that at the beginning of a syllable, we really like there to be changes in how open the mouth and the rest of the vocal tract is as we get closer to the vowel. We generally prefer this to get more open, since the mouth is basically totally open when we pronounce a vowel, and so as we go through, we want to get ready for the vowel-ing. But we really don't like it when we trade from one sound to another of the same type.

That's why we don't just think words starting with [sf] are bad, like "sfeek," but we'd probably also see "stheek", "s-sheek", and "sveek" pretty bad as well. (For you IPA lovers out there, I'm aiming for [sθik], [sʃik], and [svik].) All of those sounds have the same method of pronouncing them, where just a little air can escape through the vocal tract. And ones where the airflow's totally stopped? We don't like starting off with two of those, either: "pteek", "kdeek", and "tgeek" also probably sound pretty terrible.

As for our other example: when we have a sentence like "Dirk and Jake expect to see them in dreams," we may not know who "them" is pointing to, but we do know it's not Dirk and Jake. But if we morph that into the question "Who do Dirk and Jake expect to see them in dreams?", suddenly, we can have Dirk and Jake be the same people as "them." What's going on there?

We can't really give a solid explanation about this without really delving into syntax and the relationships between pronouns and the nouns they can latch onto in a sentence. That's a whole other future episode. But we can talk about this in a way where we can look at rules for interpretation.

So we want to start by looking at the second part of each of these sentences. Each of them have a second, infinitive verb in their latter half: "to see them in dreams." But in both cases, there's no immediately apparent subject in this sentence: we don't know from just the words that are there who exactly is doing the seeing. It's just not there.

But we do actually get an interpretation for each of these for who is doing the seeing. In "Dirk and Jake expect to see them in dreams," the people doing the seeing are Dirk and Jake. However, in the question version, "Who do Dirk and Jake expect to see them in dreams?", suddenly, the people doing the seeing is whoever the answer to the question is. It's "who!" "Who" is doing the seeing.

So apparently, there's a restriction on interpretation for pronouns, then. When it's the main subject in the same clause, you can't have the subject be the same person or people as the pronoun. When it's further away, though, like the subject of a different clause, then it's totally fine.

That's an approximation of the answer, but it gives you an idea of what it looks like - and again, it shows there's a rule here you've probably never thought about that tells you how you can interpret these sentences. Pretty cool!



So how about it? What do you all think? Let us know below, and we’ll be happy to talk with you about prescriptivism vs. descriptivism, or your favorite arguments around both kinds of rules. There’s a lot of interesting stuff to say, and so we want to hear what interests you!

blog comments powered by Disqus

Previous Topic:                                                                                                                                                                                                            Next Topic:

The Rules of Conversation                                                                                                                 Now You Hear It, Now You Don't, Part 1: Phonemes